Distributed data retrieval system including mechanism for predicting an amount of response data

• US 6,633,873 B1
• Filed: 08/17/1999
• Issued: 10/14/2003
• Est. Priority Date: 02/16/1999
• Status: Expired due to Fees

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Nakamura, Naohito
• Primary Examiner(s): Homere, Jean R.
• Assistant Examiner(s): Wassum, Luke S
• Application Number:
  US09/375,388
• Time in Patent Office: 1,519 Days
• Field of Search: 707/10, 707/3, 709/203, 709/219
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2471 Distributed queries
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

When a predetermined number of pieces of data are retrieved, a retrieving process is performed first on a local database 7 in a LAN 1, and a response counter 8 counts the number of pieces of response data accumulated in a response accumulation device 9. When it is determined that the predetermined number of pieces of response data cannot be obtained from the local database, a remote database 3 is searched by a retrieval device 6 through a communications device 2. Therefore, requests to search unnecessary databases can be reduced, and the load on a network and the remote database 3 can also be reduced successfully.

Claims

1. A data access apparatus functioning as a distributed retrieval system in which a plurality of databases are distributed in a network, and retrieving response data satisfying a predetermined condition based on a specific database, comprising:

a retrieval unit retrieving over the network from the specific database, and obtaining data satisfying the predetermined condition as a response;

a count unit generating count information by counting a number of pieces of response data as they are being obtained over the network by said retrieval unit searching the specific database;

a prediction unit predicting a likely number of pieces of response data to be finally obtained over the network as a result of said retrieval unit searching the specific database according to the count information from said count unit and information about the amount of data stored in the specific database; and

said retrieval unit requesting over the network to search another database in the network when a number of pieces of prediction responses obtained by said prediction unit does not reach a predetermined value, where the search of the other database occurs while the data from the specific database is being obtained.

2. The apparatus according to claim 1, wherein

while said retrieval unit is searching the specific database, said prediction unit obtains the predicted likely number of responses, which is proportional to the number of pieces of retrieved response data indicated by the count information and the number of records to be retrieved, relative to the number of retrieved records, and

according to information about a number of pieces of retrieved response data counted by said count unit, a number of records to be retrieved in the specific database, and information about a number of retrieved records of said retrieval unit.

3. The apparatus according to claim 1, wherein said retrieval unit concurrently searches a sub-plurality of other databases in the network, and further comprising:

a retrieval stop request unit transmitting over the network a request to stop searching to the other databases when a predetermined number of pieces of response data can be obtained after issuing a retrieval request from said retrieval request unit over the network to the other databases.

4. An apparatus according to claim 1, further comprising:

a second prediction unit predicting a number of pieces of response data likely to be obtained from each of the other databases, wherein

said retrieval unit determines the particular number of pieces of response data requested over the network from each of the other databases using a prediction result of said second prediction unit.

5. An apparatus according to claim 2, wherein said retrieval unit determines the particular number of pieces of response data requested over the network from each of the other databases using a prediction result of said prediction unit.

6. A data access apparatus functioning as a distributed retrieval system in which a plurality of databases are distributed in a network, and retrieving response data satisfying a predetermined condition based on a specific database, comprising:

a retrieval unit retrieving over the network from the specific database and obtaining data satisfying the predetermined condition as a response, requesting no more than a particular number of pieces of appropriately retrieved response data from each of other databases in the network when a retrieval request message is issued over the network to the other databases, and receiving over the network response data from the other databases obtained as a result of searching each of the other databases responsive to the request message issued from said retrieval unit; and

a prediction unit predicting a number of pieces of response data likely to be obtained from each of the other databases, wherein

said retrieval unit determines the particular number of pieces of response data requested over the network from each of the other databases using a prediction result of said prediction unit.

7. A data accessing method functioning as a distributed retrieval system in which a plurality of databases are distributed in a network, and retrieving response data satisfying a predetermined condition based on a specific database, comprising:

retrieving to a search system from the specific database, and obtaining over the network data satisfying the predetermined condition as a response;

counting a number of pieces of response data obtained by said retrieving the specific database;

predicting at the search system a number of pieces of response data to be finally obtained as a result of said retrieving according to count information from said counting and according to information about an amount of data stored in the specific database; and

retrieval-requesting over the network to search another database in the network when a number of pieces of prediction responses obtained in said predicting does not reach a predetermined value.

8. A method according to claim 7, where the specific database is automatically selected based on whether it is on a local or a remote network.

9. A computer-readable storage medium, in a system functioning as a distributed retrieval system in which a plurality of databases are distributed in a network, and retrieving response data satisfying a predetermined condition based on a specific database, storing a program used to direct a computer to perform the functions of:

retrieving to a search system from the specific database, and obtaining over the network data satisfying the predetermined condition as a response to a message sent by the search system;

counting at the search system a number of pieces of response data obtained by said retrieving function searching the specific database;

predicting at the search system a number of pieces of response data to be finally obtained as a result of said retrieving function according to count information from said counting function and according to information about an amount of data stored in the specific database; and

retrieval-requesting to search another database in the network when a number of pieces of prediction responses obtained in said predicting function does not reach a predetermined value by the search system sending a retrieval request message over the network to the other database.